RESEARCH ARTICLE
Comparative Study of Extracellular Vesicles from the Urine of Healthy Individuals and Prostate Cancer Patients Olga E. Bryzgunova1, Marat M. Zaripov2, Tatyana E. Skvortsova1, Evgeny A. Lekchnov1, Alina E. Grigor’eva3, Ivan A. Zaporozhchenko1*, Evgeny S. Morozkin1,4, Elena I. Ryabchikova3, Yuri B. Yurchenko5, Vladimir E. Voitsitskiy2, Pavel P. Laktionov1,4
a11111
1 Laboratory of molecular medicine, Institute of Chemical Biology and Fundamental Medicine CD RAS, Novosibirsk, Russian Federation, 2 Dispensary department 2, Novosibirsk Regional Oncology Center, Novosibirsk, Russian Federation, 3 Group of microscopy, Institute of Chemical Biology and Fundamental Medicine CD RAS, Novosibirsk, Russian Federation, 4 Centre of Oncology and Radiotherapy, Novosibirsk Research Institute of Circulation Pathology Academician E.N. Meshalkin, Novosibirsk, Russian Federation, 5 Center of New Medical Technologies of ICBFM SB RAS, Novosibirsk, Russian Federation * [email protected]
Abstract OPEN ACCESS Citation: Bryzgunova OE, Zaripov MM, Skvortsova TE, Lekchnov EA, Grigor’eva AE, Zaporozhchenko IA, et al. (2016) Comparative Study of Extracellular Vesicles from the Urine of Healthy Individuals and Prostate Cancer Patients. PLoS ONE 11(6): e0157566. doi:10.1371/journal.pone.0157566 Editor: David Raul Francisco Carter, Oxford Brookes University, UNITED KINGDOM Received: February 23, 2016 Accepted: June 1, 2016 Published: June 15, 2016 Copyright: © 2016 Bryzgunova et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This work was supported by Russian Science Foundation (project № 16-15-00124). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Recent studies suggest that extracellular vesicles may be the key to timely diagnosis and monitoring of genito-urological malignancies. In this study we investigated the composition and content of extracellular vesicles found in the urine of healthy donors and prostate cancer patients. Urine of 14 PCa patients and 20 healthy volunteers was clarified by low-speed centrifugation and total extracellular vesicles fraction was obtain by high-speed centrifugation. The exosome-enriched fraction was obtained by filtration of total extracellular vesicles through a 0.1 μm pore filter. Transmission electron microscopy showed that cell-free urine in both groups contained vesicles from 20 to 230 nm. Immunogold staining after ultrafiltration demonstrated that 95% and 90% of extracellular vesicles in healthy individuals and cancer patients, respectively, were exosomes. Protein, DNA and RNA concentrations as well as size distribution of extracellular vesicles in both fractions were analyzed. Only 75% of the total protein content of extracellular vesicles was associated with exosomes which amounted to 90–95% of all vesicles. Median DNA concentrations in total extracellular vesicles and exosome-enriched fractions were 18 pg/ml and 2.6 pg/ml urine, correspondingly. Urine extracellular vesicles carried a population of RNA molecules 25 nt to 200 nt in concentration of no more than 290 pg/ml of urine. Additionally, concentrations of miR-19b, miR-25, miR-125b, and miR-205 were quantified by qRT-PCR. MiRNAs were shown to be differently distributed between different fractions of extracellular vesicles. Detection of miR-19b versus miR-16 in total vesicles and exosome-enriched fractions achieved 100%/93% and 95%/ 79% specificity/sensitivity in distinguishing cancer patients from healthy individuals, respectively, demonstrating the diagnostic value of urine extracellular vesicles.
Competing Interests: The authors have declared that no competing interests exist.
PLOS ONE | DOI:10.1371/journal.pone.0157566 June 15, 2016
1 / 17
Urine Extracellular Vesicles in Prostate Cancer
Introduction Prostate cancer (PCa) is the second most common cancer worldwide in males, with more than 1.1 million new cases diagnosed in 2012 (global cancer statistics, http://www.cancerresearchuk. org/). Despite five-year survival rate reaching 98% in developed countries, early PCa detection and accurate post-therapy monitoring for tumor recurrence, proliferation and metastatic potential is demanded. It can increase the quality of life for PCa patients, ensure timely diagnosis and survival of patients diagnosed at an advance stage. Despite a number of shortcomings and U.S. Preventive Services Task Force recommendation against its use, blood PSA is still used for PCa diagnostics [1]. Men with a high PSA are required to undergo additional tests such digital rectal exam or prostate biopsy, which are both uncomfortable and may cause adverse after-effects i.e. a needle biopsy can result in infection or prolonged bleeding afterwards. PCA3 assay despite very good initial performance [2,3] was later shown to possess low sensitivity and specificity (69 and 58%, correspondingly) [4]. Thus, a non-invasive test for PCa is still desired. The prostate ejaculatory ducts empty directly into the urethra, carrying the prostate secretions into the urinary tract. Thus, urine represents a potentially valuable source of diagnostic material for monitoring the prostate. Indeed, it has been shown that cell-free DNA from the urine can be used for PCa diagnostics, and simple procedures like prostate massage can increase the amount of tumor-specific nucleic acids in urine and subsequently the efficacy of PCa diagnostics [5,6]. The low concentration of tumor-specific molecules demands a special protocol for their isolation from large urine volumes as well as a highly sensitive quantification assay. This seemingly decreases the attractiveness of urine as a source of diagnostic material. Recently, however, certain types of extracellular vesicles (EVs), enriched in biopolymers originating from cancer cells were found in urine [7–9]. The most interesting are exosomes, a subclass of extracellular vesicles ~ 30–150 nm in diameter, containing a portion of the parent cell cytoplasm [10]. Exosomes are released into the extracellular space after merging of multivesicular bodies with the cell membrane and are subsequently passed into the blood, urine and other biological fluids. In contrast, microvesicles are formed from the plasma membrane, and are more heterogeneous in size [11,12]. Both microvesicles and exosomes have been shown to contain a snapshot of the nucleic acid content of the parent cell [13]. A comprehensive analysis of the protein content of EVs found in urine showed the presence of proteins/transporters specific to cells of the kidney and urogenital tract [14,15]. Later, it was shown that sufficiently stable urine microvesicles carry miRNA, and also have small amounts of DNA at their surface [13] and, similar to blood EVs, have the potential to be used as a source of biomarkers for the detection of genitourinary pathologies [16]. There are examples of transcriptomics and proteomics studies of urinary EVs. Royo and colleagues performed transcriptomic profiling of urinary EVs obtained from prostate cancer and benign prostate hyperplasia patients using HumanHT-12 v4 Expression BeadChip and found two RNA transcripts, Cadherin 3, type 1 (CDH3) and CKLF-Like MARVEL Transmembrane Domain Containing 3 (CMTM3), exhibited the predicted behavior [17]. Overbye and colleagues reported a mass spectroscopy proteomic study of urinary exosomes in order to identify proteins differentially expressed in PCa patients and healthy male controls [18]. In the separate studies urine EVs were studied in respect to their size [19,20], protein, RNA, DNA and miRNA content [21,22]. Differences in EV isolation including different centrifugation regimes and filtration as well as use of different size measurement methodologies lead to the discrepancy in the results concealing diagnostic significance of different subclasses of EVs. It is still not clearly established what particles are enriched in protein or nucleic acids, and if there are morphological differences in EVs in health and disease.
PLOS ONE | DOI:10.1371/journal.pone.0157566 June 15, 2016
2 / 17
Urine Extracellular Vesicles in Prostate Cancer
Here we investigated EVs from cell-free urine obtained by different centrifugation and filtration protocols. Morphology, size distribution, protein, DNA and RNA content, as well as prostate cancer-specific miRNA markers were studied in cell-free urine fractions including total EV (TEV) and small EV (ERV) (below 100 nm). Special attention was given to the comparison of EVs from healthy individuals and prostate cancer patients and utility of EVs for prostate cancer diagnostics.
Materials and Methods Study population and blood/urine collection Blood (used only to determine the PSA level) and urine samples of 20 healthy male individuals (HD) and 14 previously untreated prostate cancer patients (PCa) were obtained from Center of New Medical technologies of ICBFM SB RAS and Regional Oncology Center (Novosibirsk, Russia) (Table 1). None of the patients had undergone surgical treatment or received chemotherapy prior to/at the time of sample collection. The work was conducted in compliance with the principles of voluntariness and confidentiality in accordance with the “Fundamentals of Legislation on Health Care”. The study was approved by the ethics committees of ICBFM SB RAS and Novosibirsk Regional Oncology Center and written informed consent was provided by all participants.
Urine fractionation and isolation of extracellular vesicles To pellet cells, fresh urine was clarified by centrifugation at 400g, 20°C, for 20 min within 3 hours after collection. Supernatants were then centrifuged at 17000g, 20°C, for 20 min. Aliquots from both supernatants were immediately frozen and stored at -20°C until use. Aliquots were thawed once. Total EVs (TEV) were precipitated from the 17 000g supernatant by high-speed centrifugation at 100000g, 18°C, for 90 min, the pellet was washed with 10 ml PBS and re-suspended in 100–300 μl of PBS. To obtain the exosome-enriched fraction of vesicles (ERV) TEV fraction was size selected after first centrifugation by syringe filtration through a 0.1 μm pore filter (Minisort High-Flow, Sartorius) and precipitated by high speed centrifugation. The pellets were resuspended in 100–300 μl of PBS and immediately used for transmission electron Table 1. Overview of the study population. HD (n = 20)
PCa (n = 14)
Mean±SEM
59±1.7
72±1.5
Range
Age 48–73
63–82
Total PSA, ng/ml
1.1±0.15
20±3
PCa stage
N/A
T2-3NXMX,1
T2
N/A
7 (50%)
T3
N/A
7 (50%)
NX
N/A
14 (100%)
MX
N/A
13 (93%)
M1
N/A
1 (7%)
Gleason score 5–6
N/A
1 (7%)
Gleason score 6
N/A
5 (36%)
Gleason score 7
N/A
8 (57%)
doi:10.1371/journal.pone.0157566.t001
PLOS ONE | DOI:10.1371/journal.pone.0157566 June 15, 2016
3 / 17
Urine Extracellular Vesicles in Prostate Cancer
microscopy observations. Preparations of EVs in PBS were snap frozen in liquid nitrogen and stored at -80°C until use. Aliquots were thawed once immediately before use. Where noted concentrations of the biopolymers were normalized to the initial urine volume according to dilution ratios calculated for each faction.
Transmission electron microscopy Fresh samples of extracellular vesicles (20 μl) were adsorbed for 1 min on copper grids covered with formvar film and stabilized by carbon. The grids were exposed for 5–10 sec on a drop of 0.5% uranyl acetate, then the excess of fluid was removed using filter paper and the grids were air dried. The size distribution of vesicles was evaluated using 36 TEM images from three healthy donors and three PCa patients (six images per patient, a median of ~39 vesicles per image, no less than 200 per patient). Counting was performed manually, with increments of 10 nm; for example, the 31–40 nm group contains vesicles sized 30.1 nm, but 40 nm. To evaluate the expression of surface marker proteins, 10 μl of thawed suspension was incubated for 18 h on an orbital shaker with 10 μl of 0.5% BSA/PB (phosphate buffer) and 3 μl of mouse monoclonal antiCD63, CD24 and CD9 antibodies (Abcam, 100 μg/ml). After incubation, vesicles were adsorbed for 1 min on copper grids covered with formvar film and stabilized by carbon, rinsed twice with PBS, and incubated with a protein A-colloidal gold conjugate for 2 h in a humid chamber on a shaker at room temperature followed by two washes with PBS, then stained with phosphotungstic acid, as described previously [23,24]. Grids were analyzed using a JEM 1400 (80 kV, Jeol, Japan) transmission electron microscope supplied with a digital camera Veleta (Olympus SIS, Germany). The measurements were performed using iTEM (Olympus SIS, Germany) software.
Protein measurements Protein concentrations were measured using the NanoOrange Protein Quantitation Kit (Invitrogen, USA) according to the manufacturer's protocol. Preparations of EVs were lysed for 10 min on ice in lysis buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 0.1 M DTT), incubated at 95°C for 10 min with a working solution of NanoOrange reagent. Fluorescence was measured on a VersaFluor™ fluorometer at 480 nm excitation and 580 nm emission filters against standard solutions of BSA.
Nucleic acid isolation and analysis Total DNA and RNA from the total microvesicles and exosomes were isolated using a DNA or RNA isolation kit (Biosilica Ltd, Russia), followed by precipitation of the nucleic acids using acetone and triethylamine, as described earlier [25]. DNA concentration was measured using multiplex TaqMan PCR (α-satellite elements and LINE1 repeats) [25]. RNA samples were treated with DNase I (Fermentas) at 37°C for 30 min, and one RNA sample from a PCa patient was further treated with RNase A (Fermentas). The RNA size distribution was analyzed using an Agilent 2100 Bioanalyzer capillary electrophoresis system and RNA 6000 Pico Kit. MiRNA from all fractions were isolated using single-phase protocol as described [26]. Concentrations of miRNA (miR-16, miR-19b, miR-25, miR-125b, and miR-205) were determined by TaqMan PCR after reverse transcription with stem-loop primers, as described earlier [26]. Primers and probes for reverse transcription and quantitative PCR are listed in S1 Table.
Statistical analysis Statistical analyses were performed using GraphPad Prism 5 software. The specificity and sensitivity of the analytical systems was characterized by Receiving Operator Characteristic curves (ROC-curves). Threshold cycle values (Ct) values for all miRNAs obtained in qRT-PCR were
PLOS ONE | DOI:10.1371/journal.pone.0157566 June 15, 2016
4 / 17
Urine Extracellular Vesicles in Prostate Cancer
normalized to miR-16 to obtain delta Ct values (dCt). Normalizing was done by subtracting Ct values for miR-16 from Ct values for target miRNA for each sample.
Results The isolation protocol used in this study yielded four fractions. First (400g supernatant) fraction is clarified cell-free urine and presumably contains all types of vesicles present in the urine as well as urine proteins, protein aggregates, cell fragments and debris, and cell-free nucleic acids not associated with membrane vesicles. Second fraction (17 000g supernatant) contains urine EVs, soluble urine protein and cell-free nucleic acids not packed in EVs. Third fraction containing total extracellular vesicles (TEV) is a pellet of high-speed centrifugation and is comprised of exosomes and smaller microvesicles (up to 200–300 nm). Finally, the exosome-rich fraction of vesicles (ERV) obtained by 0.1 μm filtration of TEV fraction is enriched in smaller vesicles, including exosomes. The composition of EVs was studied in TEV and ERV fractions, except miRNA concentration, which was measured in all fractions.
Description of urine extracellular vesicles According to TEM, TEV from urine of healthy donors and PCa patients contained EVs sized from 20 to 230 nm (Fig 1A). The majority of vesicles had the appearance of spherical bubbles or “cups” hinting at their exosomal/endocytic origin. Vesicles ranging 30–100 nm (exosome size range) amounted to 95% and 90% of the total number of vesicles in TEV samples from HD and PCa respectively; this difference was not significant (Mann-Whitney test) (Fig 1B). Filtration of TEV fraction through a 0.1 μm filter increased the portion of 30–100 nm exosome-like vesicles up to almost 100%. The vesicles exceeding 110 nm were almost completely eliminated by 0.1 μm filtration (Fig 1B). Immunogold staining of 0.1 μm filtered samples with antibodies against CD63, CD9 and CD24 was positive for virtually all 30 to 100 nm vesicles, suggesting their exosomal origin (Fig 1A).
Quantification of total protein in urine EVs Indirectly, the number of microvesicles and exosomes can be estimated from the total protein concentration in their preparations. In our study protein concentration in preparations of urine EVs was measured by NanoOrange Protein Quantification Kit, which includes heating of sample with detergent to lyse the vesicles. The protein concentration in the preparations of urine EVs did not differ between cancer patients and healthy individuals (Fig 2). The concentration of protein in preparations of urine EVs was 12±0.7 mg/ml and 9±0.4 mg/ml, respectively (Fig 2A), which corresponds to 171±22 ng and 140±21 ng per ml of urine used (Fig 2B). Considering the normal concentration of urine protein (about 33 μg /ml) and the dilution during the EV isolation including resuspension and centrifugation steps, EV preparations contained no more than 3 ng/ml of remaining urine protein. Thus protein content in preparations of EVs amounts to approximately 0.6% of total urine protein. It should be noted that only 75% of the total protein of the TEV was associated with ERV, which constitute 90–95% of all vesicles. Thus, 5–10% of large vesicles (>110 nm) contained 25% of the total protein associated with EV, demonstrating the different nature of small (100
Comparative Study of Extracellular Vesicles from the Urine of Healthy Individuals and Prostate Cancer Patients Olga E. Bryzgunova1, Marat M. Zaripov2, Tatyana E. Skvortsova1, Evgeny A. Lekchnov1, Alina E. Grigor’eva3, Ivan A. Zaporozhchenko1*, Evgeny S. Morozkin1,4, Elena I. Ryabchikova3, Yuri B. Yurchenko5, Vladimir E. Voitsitskiy2, Pavel P. Laktionov1,4
a11111
1 Laboratory of molecular medicine, Institute of Chemical Biology and Fundamental Medicine CD RAS, Novosibirsk, Russian Federation, 2 Dispensary department 2, Novosibirsk Regional Oncology Center, Novosibirsk, Russian Federation, 3 Group of microscopy, Institute of Chemical Biology and Fundamental Medicine CD RAS, Novosibirsk, Russian Federation, 4 Centre of Oncology and Radiotherapy, Novosibirsk Research Institute of Circulation Pathology Academician E.N. Meshalkin, Novosibirsk, Russian Federation, 5 Center of New Medical Technologies of ICBFM SB RAS, Novosibirsk, Russian Federation * [email protected]
Abstract OPEN ACCESS Citation: Bryzgunova OE, Zaripov MM, Skvortsova TE, Lekchnov EA, Grigor’eva AE, Zaporozhchenko IA, et al. (2016) Comparative Study of Extracellular Vesicles from the Urine of Healthy Individuals and Prostate Cancer Patients. PLoS ONE 11(6): e0157566. doi:10.1371/journal.pone.0157566 Editor: David Raul Francisco Carter, Oxford Brookes University, UNITED KINGDOM Received: February 23, 2016 Accepted: June 1, 2016 Published: June 15, 2016 Copyright: © 2016 Bryzgunova et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: This work was supported by Russian Science Foundation (project № 16-15-00124). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Recent studies suggest that extracellular vesicles may be the key to timely diagnosis and monitoring of genito-urological malignancies. In this study we investigated the composition and content of extracellular vesicles found in the urine of healthy donors and prostate cancer patients. Urine of 14 PCa patients and 20 healthy volunteers was clarified by low-speed centrifugation and total extracellular vesicles fraction was obtain by high-speed centrifugation. The exosome-enriched fraction was obtained by filtration of total extracellular vesicles through a 0.1 μm pore filter. Transmission electron microscopy showed that cell-free urine in both groups contained vesicles from 20 to 230 nm. Immunogold staining after ultrafiltration demonstrated that 95% and 90% of extracellular vesicles in healthy individuals and cancer patients, respectively, were exosomes. Protein, DNA and RNA concentrations as well as size distribution of extracellular vesicles in both fractions were analyzed. Only 75% of the total protein content of extracellular vesicles was associated with exosomes which amounted to 90–95% of all vesicles. Median DNA concentrations in total extracellular vesicles and exosome-enriched fractions were 18 pg/ml and 2.6 pg/ml urine, correspondingly. Urine extracellular vesicles carried a population of RNA molecules 25 nt to 200 nt in concentration of no more than 290 pg/ml of urine. Additionally, concentrations of miR-19b, miR-25, miR-125b, and miR-205 were quantified by qRT-PCR. MiRNAs were shown to be differently distributed between different fractions of extracellular vesicles. Detection of miR-19b versus miR-16 in total vesicles and exosome-enriched fractions achieved 100%/93% and 95%/ 79% specificity/sensitivity in distinguishing cancer patients from healthy individuals, respectively, demonstrating the diagnostic value of urine extracellular vesicles.
Competing Interests: The authors have declared that no competing interests exist.
PLOS ONE | DOI:10.1371/journal.pone.0157566 June 15, 2016
1 / 17
Urine Extracellular Vesicles in Prostate Cancer
Introduction Prostate cancer (PCa) is the second most common cancer worldwide in males, with more than 1.1 million new cases diagnosed in 2012 (global cancer statistics, http://www.cancerresearchuk. org/). Despite five-year survival rate reaching 98% in developed countries, early PCa detection and accurate post-therapy monitoring for tumor recurrence, proliferation and metastatic potential is demanded. It can increase the quality of life for PCa patients, ensure timely diagnosis and survival of patients diagnosed at an advance stage. Despite a number of shortcomings and U.S. Preventive Services Task Force recommendation against its use, blood PSA is still used for PCa diagnostics [1]. Men with a high PSA are required to undergo additional tests such digital rectal exam or prostate biopsy, which are both uncomfortable and may cause adverse after-effects i.e. a needle biopsy can result in infection or prolonged bleeding afterwards. PCA3 assay despite very good initial performance [2,3] was later shown to possess low sensitivity and specificity (69 and 58%, correspondingly) [4]. Thus, a non-invasive test for PCa is still desired. The prostate ejaculatory ducts empty directly into the urethra, carrying the prostate secretions into the urinary tract. Thus, urine represents a potentially valuable source of diagnostic material for monitoring the prostate. Indeed, it has been shown that cell-free DNA from the urine can be used for PCa diagnostics, and simple procedures like prostate massage can increase the amount of tumor-specific nucleic acids in urine and subsequently the efficacy of PCa diagnostics [5,6]. The low concentration of tumor-specific molecules demands a special protocol for their isolation from large urine volumes as well as a highly sensitive quantification assay. This seemingly decreases the attractiveness of urine as a source of diagnostic material. Recently, however, certain types of extracellular vesicles (EVs), enriched in biopolymers originating from cancer cells were found in urine [7–9]. The most interesting are exosomes, a subclass of extracellular vesicles ~ 30–150 nm in diameter, containing a portion of the parent cell cytoplasm [10]. Exosomes are released into the extracellular space after merging of multivesicular bodies with the cell membrane and are subsequently passed into the blood, urine and other biological fluids. In contrast, microvesicles are formed from the plasma membrane, and are more heterogeneous in size [11,12]. Both microvesicles and exosomes have been shown to contain a snapshot of the nucleic acid content of the parent cell [13]. A comprehensive analysis of the protein content of EVs found in urine showed the presence of proteins/transporters specific to cells of the kidney and urogenital tract [14,15]. Later, it was shown that sufficiently stable urine microvesicles carry miRNA, and also have small amounts of DNA at their surface [13] and, similar to blood EVs, have the potential to be used as a source of biomarkers for the detection of genitourinary pathologies [16]. There are examples of transcriptomics and proteomics studies of urinary EVs. Royo and colleagues performed transcriptomic profiling of urinary EVs obtained from prostate cancer and benign prostate hyperplasia patients using HumanHT-12 v4 Expression BeadChip and found two RNA transcripts, Cadherin 3, type 1 (CDH3) and CKLF-Like MARVEL Transmembrane Domain Containing 3 (CMTM3), exhibited the predicted behavior [17]. Overbye and colleagues reported a mass spectroscopy proteomic study of urinary exosomes in order to identify proteins differentially expressed in PCa patients and healthy male controls [18]. In the separate studies urine EVs were studied in respect to their size [19,20], protein, RNA, DNA and miRNA content [21,22]. Differences in EV isolation including different centrifugation regimes and filtration as well as use of different size measurement methodologies lead to the discrepancy in the results concealing diagnostic significance of different subclasses of EVs. It is still not clearly established what particles are enriched in protein or nucleic acids, and if there are morphological differences in EVs in health and disease.
PLOS ONE | DOI:10.1371/journal.pone.0157566 June 15, 2016
2 / 17
Urine Extracellular Vesicles in Prostate Cancer
Here we investigated EVs from cell-free urine obtained by different centrifugation and filtration protocols. Morphology, size distribution, protein, DNA and RNA content, as well as prostate cancer-specific miRNA markers were studied in cell-free urine fractions including total EV (TEV) and small EV (ERV) (below 100 nm). Special attention was given to the comparison of EVs from healthy individuals and prostate cancer patients and utility of EVs for prostate cancer diagnostics.
Materials and Methods Study population and blood/urine collection Blood (used only to determine the PSA level) and urine samples of 20 healthy male individuals (HD) and 14 previously untreated prostate cancer patients (PCa) were obtained from Center of New Medical technologies of ICBFM SB RAS and Regional Oncology Center (Novosibirsk, Russia) (Table 1). None of the patients had undergone surgical treatment or received chemotherapy prior to/at the time of sample collection. The work was conducted in compliance with the principles of voluntariness and confidentiality in accordance with the “Fundamentals of Legislation on Health Care”. The study was approved by the ethics committees of ICBFM SB RAS and Novosibirsk Regional Oncology Center and written informed consent was provided by all participants.
Urine fractionation and isolation of extracellular vesicles To pellet cells, fresh urine was clarified by centrifugation at 400g, 20°C, for 20 min within 3 hours after collection. Supernatants were then centrifuged at 17000g, 20°C, for 20 min. Aliquots from both supernatants were immediately frozen and stored at -20°C until use. Aliquots were thawed once. Total EVs (TEV) were precipitated from the 17 000g supernatant by high-speed centrifugation at 100000g, 18°C, for 90 min, the pellet was washed with 10 ml PBS and re-suspended in 100–300 μl of PBS. To obtain the exosome-enriched fraction of vesicles (ERV) TEV fraction was size selected after first centrifugation by syringe filtration through a 0.1 μm pore filter (Minisort High-Flow, Sartorius) and precipitated by high speed centrifugation. The pellets were resuspended in 100–300 μl of PBS and immediately used for transmission electron Table 1. Overview of the study population. HD (n = 20)
PCa (n = 14)
Mean±SEM
59±1.7
72±1.5
Range
Age 48–73
63–82
Total PSA, ng/ml
1.1±0.15
20±3
PCa stage
N/A
T2-3NXMX,1
T2
N/A
7 (50%)
T3
N/A
7 (50%)
NX
N/A
14 (100%)
MX
N/A
13 (93%)
M1
N/A
1 (7%)
Gleason score 5–6
N/A
1 (7%)
Gleason score 6
N/A
5 (36%)
Gleason score 7
N/A
8 (57%)
doi:10.1371/journal.pone.0157566.t001
PLOS ONE | DOI:10.1371/journal.pone.0157566 June 15, 2016
3 / 17
Urine Extracellular Vesicles in Prostate Cancer
microscopy observations. Preparations of EVs in PBS were snap frozen in liquid nitrogen and stored at -80°C until use. Aliquots were thawed once immediately before use. Where noted concentrations of the biopolymers were normalized to the initial urine volume according to dilution ratios calculated for each faction.
Transmission electron microscopy Fresh samples of extracellular vesicles (20 μl) were adsorbed for 1 min on copper grids covered with formvar film and stabilized by carbon. The grids were exposed for 5–10 sec on a drop of 0.5% uranyl acetate, then the excess of fluid was removed using filter paper and the grids were air dried. The size distribution of vesicles was evaluated using 36 TEM images from three healthy donors and three PCa patients (six images per patient, a median of ~39 vesicles per image, no less than 200 per patient). Counting was performed manually, with increments of 10 nm; for example, the 31–40 nm group contains vesicles sized 30.1 nm, but 40 nm. To evaluate the expression of surface marker proteins, 10 μl of thawed suspension was incubated for 18 h on an orbital shaker with 10 μl of 0.5% BSA/PB (phosphate buffer) and 3 μl of mouse monoclonal antiCD63, CD24 and CD9 antibodies (Abcam, 100 μg/ml). After incubation, vesicles were adsorbed for 1 min on copper grids covered with formvar film and stabilized by carbon, rinsed twice with PBS, and incubated with a protein A-colloidal gold conjugate for 2 h in a humid chamber on a shaker at room temperature followed by two washes with PBS, then stained with phosphotungstic acid, as described previously [23,24]. Grids were analyzed using a JEM 1400 (80 kV, Jeol, Japan) transmission electron microscope supplied with a digital camera Veleta (Olympus SIS, Germany). The measurements were performed using iTEM (Olympus SIS, Germany) software.
Protein measurements Protein concentrations were measured using the NanoOrange Protein Quantitation Kit (Invitrogen, USA) according to the manufacturer's protocol. Preparations of EVs were lysed for 10 min on ice in lysis buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 0.1 M DTT), incubated at 95°C for 10 min with a working solution of NanoOrange reagent. Fluorescence was measured on a VersaFluor™ fluorometer at 480 nm excitation and 580 nm emission filters against standard solutions of BSA.
Nucleic acid isolation and analysis Total DNA and RNA from the total microvesicles and exosomes were isolated using a DNA or RNA isolation kit (Biosilica Ltd, Russia), followed by precipitation of the nucleic acids using acetone and triethylamine, as described earlier [25]. DNA concentration was measured using multiplex TaqMan PCR (α-satellite elements and LINE1 repeats) [25]. RNA samples were treated with DNase I (Fermentas) at 37°C for 30 min, and one RNA sample from a PCa patient was further treated with RNase A (Fermentas). The RNA size distribution was analyzed using an Agilent 2100 Bioanalyzer capillary electrophoresis system and RNA 6000 Pico Kit. MiRNA from all fractions were isolated using single-phase protocol as described [26]. Concentrations of miRNA (miR-16, miR-19b, miR-25, miR-125b, and miR-205) were determined by TaqMan PCR after reverse transcription with stem-loop primers, as described earlier [26]. Primers and probes for reverse transcription and quantitative PCR are listed in S1 Table.
Statistical analysis Statistical analyses were performed using GraphPad Prism 5 software. The specificity and sensitivity of the analytical systems was characterized by Receiving Operator Characteristic curves (ROC-curves). Threshold cycle values (Ct) values for all miRNAs obtained in qRT-PCR were
PLOS ONE | DOI:10.1371/journal.pone.0157566 June 15, 2016
4 / 17
Urine Extracellular Vesicles in Prostate Cancer
normalized to miR-16 to obtain delta Ct values (dCt). Normalizing was done by subtracting Ct values for miR-16 from Ct values for target miRNA for each sample.
Results The isolation protocol used in this study yielded four fractions. First (400g supernatant) fraction is clarified cell-free urine and presumably contains all types of vesicles present in the urine as well as urine proteins, protein aggregates, cell fragments and debris, and cell-free nucleic acids not associated with membrane vesicles. Second fraction (17 000g supernatant) contains urine EVs, soluble urine protein and cell-free nucleic acids not packed in EVs. Third fraction containing total extracellular vesicles (TEV) is a pellet of high-speed centrifugation and is comprised of exosomes and smaller microvesicles (up to 200–300 nm). Finally, the exosome-rich fraction of vesicles (ERV) obtained by 0.1 μm filtration of TEV fraction is enriched in smaller vesicles, including exosomes. The composition of EVs was studied in TEV and ERV fractions, except miRNA concentration, which was measured in all fractions.
Description of urine extracellular vesicles According to TEM, TEV from urine of healthy donors and PCa patients contained EVs sized from 20 to 230 nm (Fig 1A). The majority of vesicles had the appearance of spherical bubbles or “cups” hinting at their exosomal/endocytic origin. Vesicles ranging 30–100 nm (exosome size range) amounted to 95% and 90% of the total number of vesicles in TEV samples from HD and PCa respectively; this difference was not significant (Mann-Whitney test) (Fig 1B). Filtration of TEV fraction through a 0.1 μm filter increased the portion of 30–100 nm exosome-like vesicles up to almost 100%. The vesicles exceeding 110 nm were almost completely eliminated by 0.1 μm filtration (Fig 1B). Immunogold staining of 0.1 μm filtered samples with antibodies against CD63, CD9 and CD24 was positive for virtually all 30 to 100 nm vesicles, suggesting their exosomal origin (Fig 1A).
Quantification of total protein in urine EVs Indirectly, the number of microvesicles and exosomes can be estimated from the total protein concentration in their preparations. In our study protein concentration in preparations of urine EVs was measured by NanoOrange Protein Quantification Kit, which includes heating of sample with detergent to lyse the vesicles. The protein concentration in the preparations of urine EVs did not differ between cancer patients and healthy individuals (Fig 2). The concentration of protein in preparations of urine EVs was 12±0.7 mg/ml and 9±0.4 mg/ml, respectively (Fig 2A), which corresponds to 171±22 ng and 140±21 ng per ml of urine used (Fig 2B). Considering the normal concentration of urine protein (about 33 μg /ml) and the dilution during the EV isolation including resuspension and centrifugation steps, EV preparations contained no more than 3 ng/ml of remaining urine protein. Thus protein content in preparations of EVs amounts to approximately 0.6% of total urine protein. It should be noted that only 75% of the total protein of the TEV was associated with ERV, which constitute 90–95% of all vesicles. Thus, 5–10% of large vesicles (>110 nm) contained 25% of the total protein associated with EV, demonstrating the different nature of small (100
